{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:VW7OKKUDXU4I2C2O64IQIT6TPO","short_pith_number":"pith:VW7OKKUD","canonical_record":{"source":{"id":"2605.13746","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T16:28:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f2c7ae9f4a961b06702d2d618e6f17bbac3408ee5b52b8231d6ef12a598d96fb","abstract_canon_sha256":"afbf576d4124999c597b4db60dd868d99d428ba28e66a06f77dbd05e9f607f96"},"schema_version":"1.0"},"canonical_sha256":"adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5","source":{"kind":"arxiv","id":"2605.13746","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13746","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13746v1","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13746","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"pith_short_12","alias_value":"VW7OKKUDXU4I","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"VW7OKKUDXU4I2C2O","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"VW7OKKUD","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:VW7OKKUDXU4I2C2O64IQIT6TPO","target":"record","payload":{"canonical_record":{"source":{"id":"2605.13746","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T16:28:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f2c7ae9f4a961b06702d2d618e6f17bbac3408ee5b52b8231d6ef12a598d96fb","abstract_canon_sha256":"afbf576d4124999c597b4db60dd868d99d428ba28e66a06f77dbd05e9f607f96"},"schema_version":"1.0"},"canonical_sha256":"adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:16.412119Z","signature_b64":"BqW5MBfSrGE7U6Wc6+oIQntXrMNhwC51TEXJt3tqQ4dGw5sM+79TXiUS/PHhp/Ga9O6FyRKXpml4OjRMq2axAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5","last_reissued_at":"2026-05-18T02:44:16.411577Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:16.411577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.13746","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7dTevA8+wEOeQwaSYfvbuTMHDy5VRLiDjlhyJOjolqDcSUV7E32kNDeidWt56y6AeWrKX6j1f49q+LoO0wezCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T11:16:53.686076Z"},"content_sha256":"6d086a82a90fb502ed100a5d09bc24ef8f93edd6fe54090d5f47d8b61c0251e6","schema_version":"1.0","event_id":"sha256:6d086a82a90fb502ed100a5d09bc24ef8f93edd6fe54090d5f47d8b61c0251e6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:VW7OKKUDXU4I2C2O64IQIT6TPO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Weakly-Supervised Spatiotemporal Anomaly Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Mubarak Shah, Praveen Tirupattur, Urvi Gianchandani","submitted_at":"2026-05-13T16:28:01Z","abstract_excerpt":"In this paper, we explore a weakly supervised method for anomaly detection. Since annotating videos is time-consuming, we only look at weak video-level labels during training. This means that given a video, we know that it is either normal or contains an anomaly, but no further annotations are used to train the network. Features are extracted from video clips that are either normal or anomalous. These features are used to determine anomaly scores for spatiotemporal regions of the clips based on a classifier and the implementation of a multiple instance ranking loss (MIL). We represent both ano"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Features extracted from normal or anomalous video clips are used to determine anomaly scores for spatiotemporal regions based on a classifier and multiple instance ranking loss, enabling detection on the UCF Crime2Local Dataset.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That video-level labels alone, combined with a standard MIL ranking loss, are sufficient to localize anomalies both spatially within frames and temporally within clips without additional supervision or post-hoc selection.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A multiple instance learning approach with ranking loss localizes spatiotemporal anomalies in videos using only video-level normal/anomalous labels on the UCF Crime2Local dataset.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a6b8c039e9924c545db9f7cae7e11d8fda284c317149874ac2b7b60a50a37753"},"source":{"id":"2605.13746","kind":"arxiv","version":1},"verdict":{"id":"c7a3a159-6ad9-41c7-9ab2-35f37fb122f2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:42:08.238408Z","strongest_claim":"Features extracted from normal or anomalous video clips are used to determine anomaly scores for spatiotemporal regions based on a classifier and multiple instance ranking loss, enabling detection on the UCF Crime2Local Dataset.","one_line_summary":"A multiple instance learning approach with ranking loss localizes spatiotemporal anomalies in videos using only video-level normal/anomalous labels on the UCF Crime2Local dataset.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That video-level labels alone, combined with a standard MIL ranking loss, are sufficient to localize anomalies both spatially within frames and temporally within clips without additional supervision or post-hoc selection.","pith_extraction_headline":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone."},"references":{"count":4,"sample":[{"doi":"","year":1900,"title":"It is made up of both normal and anomalous surveillance videos","work_id":"3436d0b8-77f5-4c8b-ab5d-4554fcadab09","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1901,"title":"Anomaly Locality in Video Surveillance","work_id":"adef97f1-af05-4a20-b059-a3c479d5d15b","ref_index":2,"cited_arxiv_id":"1901.10364","is_internal_anchor":true},{"doi":"","year":2017,"title":"A revisit of sparse coding based anomaly detection in stacked rnn framework","work_id":"68c50282-9d91-4b81-8d99-8bf2753b1d41","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Future Frame Prediction for Anomaly Detection -- A New Baseline","work_id":"e6444f49-150b-42e2-8557-fe7355c0b3ec","ref_index":4,"cited_arxiv_id":"1712.09867","is_internal_anchor":true}],"resolved_work":4,"snapshot_sha256":"04baee69210a4b541e92789fb071e0f4bc60a21a7f7f6f73b587e178b350dc69","internal_anchors":2},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"c7a3a159-6ad9-41c7-9ab2-35f37fb122f2"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jGrZJ5Zhdt1EoFfna+b/e4rLmPcDX+zAax92DeKKP25CzepgOpAdzVc460Dv7q1cJq2tmzkNGPqdn0yFqR3LAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T11:16:53.686576Z"},"content_sha256":"43d803db6cac632e4197e433dcb8bde0fb3de10171532c2465d736d3f911fec3","schema_version":"1.0","event_id":"sha256:43d803db6cac632e4197e433dcb8bde0fb3de10171532c2465d736d3f911fec3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/bundle.json","state_url":"https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-20T11:16:53Z","links":{"resolver":"https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO","bundle":"https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/bundle.json","state":"https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VW7OKKUDXU4I2C2O64IQIT6TPO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VW7OKKUDXU4I2C2O64IQIT6TPO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"afbf576d4124999c597b4db60dd868d99d428ba28e66a06f77dbd05e9f607f96","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T16:28:01Z","title_canon_sha256":"f2c7ae9f4a961b06702d2d618e6f17bbac3408ee5b52b8231d6ef12a598d96fb"},"schema_version":"1.0","source":{"id":"2605.13746","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13746","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13746v1","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13746","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"pith_short_12","alias_value":"VW7OKKUDXU4I","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"VW7OKKUDXU4I2C2O","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"VW7OKKUD","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:43d803db6cac632e4197e433dcb8bde0fb3de10171532c2465d736d3f911fec3","target":"graph","created_at":"2026-05-18T02:44:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Features extracted from normal or anomalous video clips are used to determine anomaly scores for spatiotemporal regions based on a classifier and multiple instance ranking loss, enabling detection on the UCF Crime2Local Dataset."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That video-level labels alone, combined with a standard MIL ranking loss, are sufficient to localize anomalies both spatially within frames and temporally within clips without additional supervision or post-hoc selection."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A multiple instance learning approach with ranking loss localizes spatiotemporal anomalies in videos using only video-level normal/anomalous labels on the UCF Crime2Local dataset."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone."}],"snapshot_sha256":"a6b8c039e9924c545db9f7cae7e11d8fda284c317149874ac2b7b60a50a37753"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"In this paper, we explore a weakly supervised method for anomaly detection. Since annotating videos is time-consuming, we only look at weak video-level labels during training. This means that given a video, we know that it is either normal or contains an anomaly, but no further annotations are used to train the network. Features are extracted from video clips that are either normal or anomalous. These features are used to determine anomaly scores for spatiotemporal regions of the clips based on a classifier and the implementation of a multiple instance ranking loss (MIL). We represent both ano","authors_text":"Mubarak Shah, Praveen Tirupattur, Urvi Gianchandani","cross_cats":["cs.AI"],"headline":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T16:28:01Z","title":"Weakly-Supervised Spatiotemporal Anomaly Detection"},"references":{"count":4,"internal_anchors":2,"resolved_work":4,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"It is made up of both normal and anomalous surveillance videos","work_id":"3436d0b8-77f5-4c8b-ab5d-4554fcadab09","year":1900},{"cited_arxiv_id":"1901.10364","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Anomaly Locality in Video Surveillance","work_id":"adef97f1-af05-4a20-b059-a3c479d5d15b","year":1901},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"A revisit of sparse coding based anomaly detection in stacked rnn framework","work_id":"68c50282-9d91-4b81-8d99-8bf2753b1d41","year":2017},{"cited_arxiv_id":"1712.09867","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Future Frame Prediction for Anomaly Detection -- A New Baseline","work_id":"e6444f49-150b-42e2-8557-fe7355c0b3ec","year":null}],"snapshot_sha256":"04baee69210a4b541e92789fb071e0f4bc60a21a7f7f6f73b587e178b350dc69"},"source":{"id":"2605.13746","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T19:42:08.238408Z","id":"c7a3a159-6ad9-41c7-9ab2-35f37fb122f2","model_set":{"reader":"grok-4.3"},"one_line_summary":"A multiple instance learning approach with ranking loss localizes spatiotemporal anomalies in videos using only video-level normal/anomalous labels on the UCF Crime2Local dataset.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone.","strongest_claim":"Features extracted from normal or anomalous video clips are used to determine anomaly scores for spatiotemporal regions based on a classifier and multiple instance ranking loss, enabling detection on the UCF Crime2Local Dataset.","weakest_assumption":"That video-level labels alone, combined with a standard MIL ranking loss, are sufficient to localize anomalies both spatially within frames and temporally within clips without additional supervision or post-hoc selection."}},"verdict_id":"c7a3a159-6ad9-41c7-9ab2-35f37fb122f2"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6d086a82a90fb502ed100a5d09bc24ef8f93edd6fe54090d5f47d8b61c0251e6","target":"record","created_at":"2026-05-18T02:44:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"afbf576d4124999c597b4db60dd868d99d428ba28e66a06f77dbd05e9f607f96","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T16:28:01Z","title_canon_sha256":"f2c7ae9f4a961b06702d2d618e6f17bbac3408ee5b52b8231d6ef12a598d96fb"},"schema_version":"1.0","source":{"id":"2605.13746","kind":"arxiv","version":1}},"canonical_sha256":"adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5","first_computed_at":"2026-05-18T02:44:16.411577Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:16.411577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BqW5MBfSrGE7U6Wc6+oIQntXrMNhwC51TEXJt3tqQ4dGw5sM+79TXiUS/PHhp/Ga9O6FyRKXpml4OjRMq2axAw==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:16.412119Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13746","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d086a82a90fb502ed100a5d09bc24ef8f93edd6fe54090d5f47d8b61c0251e6","sha256:43d803db6cac632e4197e433dcb8bde0fb3de10171532c2465d736d3f911fec3"],"state_sha256":"217e4b8cdad9a94f96ddef4725ac608fd076d93a422b4e5120c741fabddf9549"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M2mFxa3bOPCj7ErlIbdCH46/DbZKNPsqOyAqy04aXUkN0d4cuLum256WvsTBwz/trtkV+KVRwyr8ZAVvXD1QDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-20T11:16:53.688943Z","bundle_sha256":"e6b2dedfe390ae6e4910cf78b934f46582e370066c58239db10a13eb943aed30"}}